Many strategies for gene therapy of solid tumors have involved the use of Moloney murine leukemia (MLV)-based retroviral vectors to deliver cytotoxic of immunogenic genes or anti-oncogenes, but problems have been encountered due to lack of adequate levels of transduction, even when virus packaging cells have been injected directly into tumors. Gene transfer using replication-competent retroviral vectors would be more efficient, as each tumor cell which is successfully transduced would itself become a virus-producing cell and initiate further infection even after the initial injection. Although replication competent retroviral (RCR) vectors for delivery of non-viral genes have been described previously, one of these proved to be stable, resulting in rapid rearrangement and deletion of the transgenes, usually within one or two replication cycles. We have devised a novel construct design that has proven stable over at least 8 serial passages in cell culture, and is capable of highly efficient gene delivery to solid tumors in vivo in a nude mouse xenograft model. Until now, the use of such RCR vectors has rarely been contemplated due to their instability and risks associated with uncontrolled spread of virus. However, as we have already established the feasibility of using RCR vectors to achieve highly efficient gene delivery to tumor cells both in cell culture and in vivo, we now propose to target such vectors specifically to prostate tumors by replacing the retroviral promoter in the LTR with the prostate-specific probasin promoter in order to control transcription of the RCR vector genome and hence its replication. We will undertake basic and preclinical studies to characterize the specificity and efficiency of gene delivery, therapeutic efficacy, safety, and immunogenicity of transcriptionally targeted RCR vectors in a logical, step-wise progression starting with cell culture studies, followed subcutaneous tumor models in nude mice, then subcutaneous and orthotopic tumor models in immunocompetent animals, and finally in models of distant site metastasis. The use of replication-competent retroviral vectors takes advantage of the transduction efficiency. Targeting the retrovirus specifically and exclusively to tumor cells would limit and control the replicative process and minimize the risk to normal cells, and would represent a significant improvement in vector design.